Add common interface for queuing up the non-fatal HW errors and notify
the registered kernel drivers.The interface supports drivers to register
to receive the callback for the non-fatal HW errors, including the vendor
specific HW errors, for the recovery and supports handling the non-fatal
errors in the bottom half.
Patch set
1. add the new interface to the APEI driver for the non-fatal HW
error notification.
2. change the existing error handling for the standard errors
to use the above notification interface.
2. add driver to handle HiSilicon hip PCIe controller's errors.
Changes:
V7:
1. Add changes in the APEI driver suggested by Borislav Petkov, for
queuing up all the non-fatal HW errors to the work queue and
notify the registered kernel drivers from the bottom half using
blocking notifier, common interface for both standard and
vendor-spcific errors.
2. Fix for further feedbacks in v5 HIP PCIe error handler driver
by Bjorn Helgaas.
V6:
1. Fix few changes in the patch subject line suggested by Bjorn Helgaas.
V5:
1. Fix comments from James Morse.
1.1 Changed the notification method to use the atomic_notifier_chain.
1.2 Add the error handled status for the user space.
V4:
1. Fix for the following smatch warning in the PCIe error driver,
reported by kbuild test robot<[email protected]>:
warn: should '((((1))) << (9 + i))' be a 64 bit type?
if (err->val_bits & BIT(HISI_PCIE_LOCAL_VALID_ERR_MISC + i))
^^^ This should be BIT_ULL() because it goes up to 9 + 32.
V3:
1. Fix the comments from Bjorn Helgaas.
V2:
1. Changes in the HiSilicon PCIe controller's error handling driver
for the comments from Bjorn Helgaas.
2. Changes in the APEI interface to support reporting the vendor error
for module with multiple devices, but use the same section type.
In the error handler will use socket id/sub module id etc to distinguish
the device.
V1:
1. Fix comments from James Morse.
2. add driver to handle HiSilicon hip08 PCIe controller's errors,
which is an application of the above interface.
Shiju Jose (5):
ACPI / APEI: Add support to queuing up the non-fatal HW errors and
notify
ACPI / APEI: Add callback for memory errors to the GHES notifier
ACPI / APEI: Add callback for AER to the GHES notifier
ACPI / APEI: Add callback for ARM HW errors to the GHES notifier
ACPI / APEI: Add callback for non-standard HW errors to the GHES
notifier
Yicong Yang (1):
PCI: hip: Add handling of HiSilicon HIP PCIe controller errors
drivers/acpi/apei/ghes.c | 268 ++++++++++++++++---
drivers/pci/controller/Kconfig | 8 +
drivers/pci/controller/Makefile | 1 +
drivers/pci/controller/pcie-hisi-error.c | 323 +++++++++++++++++++++++
include/acpi/ghes.h | 28 ++
5 files changed, 595 insertions(+), 33 deletions(-)
create mode 100644 drivers/pci/controller/pcie-hisi-error.c
--
2.17.1
Add callback function for handling the memory errors to the GHES notifier.
Signed-off-by: Shiju Jose <[email protected]>
---
drivers/acpi/apei/ghes.c | 55 ++++++++++++++++++++++++++++++----------
1 file changed, 42 insertions(+), 13 deletions(-)
diff --git a/drivers/acpi/apei/ghes.c b/drivers/acpi/apei/ghes.c
index 5c0ab5422311..053c4a2ed96c 100644
--- a/drivers/acpi/apei/ghes.c
+++ b/drivers/acpi/apei/ghes.c
@@ -471,23 +471,33 @@ static void ghes_clear_estatus(struct ghes *ghes,
ghes_ack_error(ghes->generic_v2);
}
-static void ghes_handle_memory_failure(struct acpi_hest_generic_data *gdata, int sev)
+static int ghes_handle_memory_failure(struct notifier_block *nb,
+ unsigned long event, void *data)
{
#ifdef CONFIG_ACPI_APEI_MEMORY_FAILURE
unsigned long pfn;
int flags = -1;
+ int sev = event;
+ struct acpi_hest_generic_data *gdata = data;
int sec_sev = ghes_severity(gdata->error_severity);
struct cper_sec_mem_err *mem_err = acpi_hest_get_payload(gdata);
+ if (!guid_equal((guid_t *)gdata->section_type, &CPER_SEC_PLATFORM_MEM))
+ return NOTIFY_DONE;
+
+ ghes_edac_report_mem_error(sev, mem_err);
+
+ arch_apei_report_mem_error(sev, mem_err);
+
if (!(mem_err->validation_bits & CPER_MEM_VALID_PA))
- return;
+ return NOTIFY_STOP;
pfn = mem_err->physical_addr >> PAGE_SHIFT;
if (!pfn_valid(pfn)) {
pr_warn_ratelimited(FW_WARN GHES_PFX
"Invalid address in generic error data: %#llx\n",
mem_err->physical_addr);
- return;
+ return NOTIFY_STOP;
}
/* iff following two events can be handled properly by now */
@@ -500,6 +510,7 @@ static void ghes_handle_memory_failure(struct acpi_hest_generic_data *gdata, int
if (flags != -1)
memory_failure_queue(pfn, flags);
#endif
+ return NOTIFY_STOP;
}
/*
@@ -547,6 +558,22 @@ static void ghes_handle_aer(struct acpi_hest_generic_data *gdata)
#endif
}
+static struct notifier_block ghes_notifier_mem_error = {
+ .notifier_call = ghes_handle_memory_failure,
+};
+
+struct ghes_error_handler_list {
+ const char *name;
+ struct notifier_block *nb;
+};
+
+static const struct ghes_error_handler_list ghes_error_handler_list[] = {
+ {
+ .name = "ghes_notifier_mem_error",
+ .nb = &ghes_notifier_mem_error,
+ },
+};
+
static BLOCKING_NOTIFIER_HEAD(ghes_event_notify_list);
/**
@@ -630,15 +657,7 @@ static void ghes_do_proc(struct ghes *ghes,
break;
}
- if (guid_equal(sec_type, &CPER_SEC_PLATFORM_MEM)) {
- struct cper_sec_mem_err *mem_err = acpi_hest_get_payload(gdata);
-
- ghes_edac_report_mem_error(sev, mem_err);
-
- arch_apei_report_mem_error(sev, mem_err);
- ghes_handle_memory_failure(gdata, sev);
- }
- else if (guid_equal(sec_type, &CPER_SEC_PCIE)) {
+ if (guid_equal(sec_type, &CPER_SEC_PCIE)) {
ghes_handle_aer(gdata);
}
else if (guid_equal(sec_type, &CPER_SEC_PROC_ARM)) {
@@ -1431,7 +1450,7 @@ static struct platform_driver ghes_platform_driver = {
static int __init ghes_init(void)
{
- int rc;
+ int rc, i;
if (acpi_disabled)
return -ENODEV;
@@ -1473,6 +1492,16 @@ static int __init ghes_init(void)
goto err;
}
+ for (i = 0; i < ARRAY_SIZE(ghes_error_handler_list); i++) {
+ const struct ghes_error_handler_list *list =
+ &ghes_error_handler_list[i];
+ rc = ghes_register_event_notifier(list->nb);
+ if (rc) {
+ pr_warn(GHES_PFX "fail to register %s\n", list->name);
+ goto err;
+ }
+ }
+
return 0;
err:
return rc;
--
2.17.1
Add callback function for handling the AER to the GHES notifier.
Signed-off-by: Shiju Jose <[email protected]>
---
drivers/acpi/apei/ghes.c | 21 ++++++++++++++++-----
1 file changed, 16 insertions(+), 5 deletions(-)
diff --git a/drivers/acpi/apei/ghes.c b/drivers/acpi/apei/ghes.c
index 053c4a2ed96c..67ef1742fc93 100644
--- a/drivers/acpi/apei/ghes.c
+++ b/drivers/acpi/apei/ghes.c
@@ -527,11 +527,16 @@ static int ghes_handle_memory_failure(struct notifier_block *nb,
* GHES_SEV_PANIC does not make it to this handling since the kernel must
* panic.
*/
-static void ghes_handle_aer(struct acpi_hest_generic_data *gdata)
+static int ghes_handle_aer(struct notifier_block *nb, unsigned long event,
+ void *data)
{
#ifdef CONFIG_ACPI_APEI_PCIEAER
+ struct acpi_hest_generic_data *gdata = data;
struct cper_sec_pcie *pcie_err = acpi_hest_get_payload(gdata);
+ if (!guid_equal((guid_t *)gdata->section_type, &CPER_SEC_PCIE))
+ return NOTIFY_DONE;
+
if (pcie_err->validation_bits & CPER_PCIE_VALID_DEVICE_ID &&
pcie_err->validation_bits & CPER_PCIE_VALID_AER_INFO) {
unsigned int devfn;
@@ -556,12 +561,17 @@ static void ghes_handle_aer(struct acpi_hest_generic_data *gdata)
pcie_err->aer_info);
}
#endif
+ return NOTIFY_STOP;
}
static struct notifier_block ghes_notifier_mem_error = {
.notifier_call = ghes_handle_memory_failure,
};
+static struct notifier_block ghes_notifier_aer = {
+ .notifier_call = ghes_handle_aer,
+};
+
struct ghes_error_handler_list {
const char *name;
struct notifier_block *nb;
@@ -572,6 +582,10 @@ static const struct ghes_error_handler_list ghes_error_handler_list[] = {
.name = "ghes_notifier_mem_error",
.nb = &ghes_notifier_mem_error,
},
+ {
+ .name = "ghes_notifier_aer",
+ .nb = &ghes_notifier_aer,
+ },
};
static BLOCKING_NOTIFIER_HEAD(ghes_event_notify_list);
@@ -657,10 +671,7 @@ static void ghes_do_proc(struct ghes *ghes,
break;
}
- if (guid_equal(sec_type, &CPER_SEC_PCIE)) {
- ghes_handle_aer(gdata);
- }
- else if (guid_equal(sec_type, &CPER_SEC_PROC_ARM)) {
+ if (guid_equal(sec_type, &CPER_SEC_PROC_ARM)) {
struct cper_sec_proc_arm *err = acpi_hest_get_payload(gdata);
log_arm_hw_error(err);
--
2.17.1
Add callback function for handling the ARM HW errors to the GHES notifier.
Signed-off-by: Shiju Jose <[email protected]>
---
drivers/acpi/apei/ghes.c | 29 +++++++++++++++++++++++------
1 file changed, 23 insertions(+), 6 deletions(-)
diff --git a/drivers/acpi/apei/ghes.c b/drivers/acpi/apei/ghes.c
index 67ef1742fc93..3b89c7621a0d 100644
--- a/drivers/acpi/apei/ghes.c
+++ b/drivers/acpi/apei/ghes.c
@@ -564,6 +564,20 @@ static int ghes_handle_aer(struct notifier_block *nb, unsigned long event,
return NOTIFY_STOP;
}
+static int ghes_handle_arm_hw_error(struct notifier_block *nb,
+ unsigned long event, void *data)
+{
+ struct acpi_hest_generic_data *gdata = data;
+ struct cper_sec_proc_arm *err = acpi_hest_get_payload(gdata);
+
+ if (!guid_equal((guid_t *)gdata->section_type, &CPER_SEC_PROC_ARM))
+ return NOTIFY_DONE;
+
+ log_arm_hw_error(err);
+
+ return NOTIFY_STOP;
+}
+
static struct notifier_block ghes_notifier_mem_error = {
.notifier_call = ghes_handle_memory_failure,
};
@@ -572,6 +586,10 @@ static struct notifier_block ghes_notifier_aer = {
.notifier_call = ghes_handle_aer,
};
+static struct notifier_block ghes_notifier_arm_hw_error = {
+ .notifier_call = ghes_handle_arm_hw_error,
+};
+
struct ghes_error_handler_list {
const char *name;
struct notifier_block *nb;
@@ -586,6 +604,10 @@ static const struct ghes_error_handler_list ghes_error_handler_list[] = {
.name = "ghes_notifier_aer",
.nb = &ghes_notifier_aer,
},
+ {
+ .name = "ghes_notifier_arm_hw_error",
+ .nb = &ghes_notifier_arm_hw_error,
+ },
};
static BLOCKING_NOTIFIER_HEAD(ghes_event_notify_list);
@@ -670,12 +692,7 @@ static void ghes_do_proc(struct ghes *ghes,
pr_warn(GHES_PFX "ghes event queue full\n");
break;
}
-
- if (guid_equal(sec_type, &CPER_SEC_PROC_ARM)) {
- struct cper_sec_proc_arm *err = acpi_hest_get_payload(gdata);
-
- log_arm_hw_error(err);
- } else {
+ {
void *err = acpi_hest_get_payload(gdata);
log_non_standard_event(sec_type, fru_id, fru_text,
--
2.17.1
Add queuing up the non-fatal HW errors to the work queue and interface to
notify the registered kernel drivers for the error recovery.
The interface enables drivers to register the callback functions to
receive the error notification, including the vendor-specific HW errors
and supports handling the non-fatal HW errors in the bottom half.
Suggested-by: Borislav Petkov <[email protected]>
Signed-off-by: Shiju Jose <[email protected]>
---
drivers/acpi/apei/ghes.c | 127 +++++++++++++++++++++++++++++++++++++++
include/acpi/ghes.h | 28 +++++++++
2 files changed, 155 insertions(+)
diff --git a/drivers/acpi/apei/ghes.c b/drivers/acpi/apei/ghes.c
index 24c9642e8fc7..5c0ab5422311 100644
--- a/drivers/acpi/apei/ghes.c
+++ b/drivers/acpi/apei/ghes.c
@@ -33,6 +33,7 @@
#include <linux/irq_work.h>
#include <linux/llist.h>
#include <linux/genalloc.h>
+#include <linux/kfifo.h>
#include <linux/pci.h>
#include <linux/pfn.h>
#include <linux/aer.h>
@@ -63,6 +64,11 @@
#define GHES_ESTATUS_CACHES_SIZE 4
#define GHES_ESTATUS_IN_CACHE_MAX_NSEC 10000000000ULL
+
+#define GHES_EVENT_RING_SIZE 256
+#define GHES_GDATA_POOL_MIN_ALLOC_ORDER 3
+#define GHES_GDATA_POOL_MIN_SIZE 65536
+
/* Prevent too many caches are allocated because of RCU */
#define GHES_ESTATUS_CACHE_ALLOCED_MAX (GHES_ESTATUS_CACHES_SIZE * 3 / 2)
@@ -122,6 +128,19 @@ static DEFINE_MUTEX(ghes_list_mutex);
*/
static DEFINE_SPINLOCK(ghes_notify_lock_irq);
+struct ghes_event_entry {
+ struct acpi_hest_generic_data *gdata;
+ int error_severity;
+};
+
+static DEFINE_KFIFO(ghes_event_ring, struct ghes_event_entry,
+ GHES_EVENT_RING_SIZE);
+
+static DEFINE_SPINLOCK(ghes_event_ring_lock);
+
+static struct gen_pool *ghes_gdata_pool;
+static unsigned long ghes_gdata_pool_size_request;
+
static struct gen_pool *ghes_estatus_pool;
static unsigned long ghes_estatus_pool_size_request;
@@ -188,6 +207,40 @@ int ghes_estatus_pool_init(int num_ghes)
return -ENOMEM;
}
+int ghes_gdata_pool_init(void)
+{
+ unsigned long addr, len;
+ int rc;
+
+ ghes_gdata_pool = gen_pool_create(GHES_GDATA_POOL_MIN_ALLOC_ORDER, -1);
+ if (!ghes_gdata_pool)
+ return -ENOMEM;
+
+ if (ghes_gdata_pool_size_request < GHES_GDATA_POOL_MIN_SIZE)
+ ghes_gdata_pool_size_request = GHES_GDATA_POOL_MIN_SIZE;
+
+ len = ghes_gdata_pool_size_request;
+ addr = (unsigned long)vmalloc(PAGE_ALIGN(len));
+ if (!addr)
+ goto err_pool_alloc;
+
+ vmalloc_sync_mappings();
+
+ rc = gen_pool_add(ghes_gdata_pool, addr, PAGE_ALIGN(len), -1);
+ if (rc)
+ goto err_pool_add;
+
+ return 0;
+
+err_pool_add:
+ vfree((void *)addr);
+
+err_pool_alloc:
+ gen_pool_destroy(ghes_gdata_pool);
+
+ return -ENOMEM;
+}
+
static int map_gen_v2(struct ghes *ghes)
{
return apei_map_generic_address(&ghes->generic_v2->read_ack_register);
@@ -247,6 +300,10 @@ static struct ghes *ghes_new(struct acpi_hest_generic *generic)
goto err_unmap_status_addr;
}
+ ghes_gdata_pool_size_request += generic->records_to_preallocate *
+ generic->max_sections_per_record *
+ generic->max_raw_data_length;
+
return ghes;
err_unmap_status_addr:
@@ -490,6 +547,49 @@ static void ghes_handle_aer(struct acpi_hest_generic_data *gdata)
#endif
}
+static BLOCKING_NOTIFIER_HEAD(ghes_event_notify_list);
+
+/**
+ * ghes_register_event_notifier - register an event notifier
+ * for the non-fatal HW errors.
+ * @nb: pointer to the notifier_block structure of the event handler.
+ *
+ * return 0 : SUCCESS, non-zero : FAIL
+ */
+int ghes_register_event_notifier(struct notifier_block *nb)
+{
+ return blocking_notifier_chain_register(&ghes_event_notify_list, nb);
+}
+EXPORT_SYMBOL_GPL(ghes_register_event_notifier);
+
+/**
+ * ghes_unregister_event_notifier - unregister the previously
+ * registered event notifier.
+ * @nb: pointer to the notifier_block structure of the event handler.
+ */
+void ghes_unregister_event_notifier(struct notifier_block *nb)
+{
+ blocking_notifier_chain_unregister(&ghes_event_notify_list, nb);
+}
+EXPORT_SYMBOL_GPL(ghes_unregister_event_notifier);
+
+static void ghes_event_work_func(struct work_struct *work)
+{
+ struct ghes_event_entry entry;
+ u32 len;
+
+ while (kfifo_get(&ghes_event_ring, &entry)) {
+ blocking_notifier_call_chain(&ghes_event_notify_list,
+ entry.error_severity,
+ entry.gdata);
+ len = acpi_hest_get_size(entry.gdata) +
+ acpi_hest_get_error_length(entry.gdata);
+ gen_pool_free(ghes_gdata_pool, (unsigned long)entry.gdata, len);
+ }
+}
+
+static DECLARE_WORK(ghes_event_work, ghes_event_work_func);
+
static void ghes_do_proc(struct ghes *ghes,
const struct acpi_hest_generic_status *estatus)
{
@@ -498,6 +598,8 @@ static void ghes_do_proc(struct ghes *ghes,
guid_t *sec_type;
const guid_t *fru_id = &guid_null;
char *fru_text = "";
+ struct ghes_event_entry event_entry;
+ u32 len;
sev = ghes_severity(estatus->error_severity);
apei_estatus_for_each_section(estatus, gdata) {
@@ -509,6 +611,25 @@ static void ghes_do_proc(struct ghes *ghes,
if (gdata->validation_bits & CPER_SEC_VALID_FRU_TEXT)
fru_text = gdata->fru_text;
+ len = acpi_hest_get_record_size(gdata);
+ event_entry.gdata = (void *)gen_pool_alloc(ghes_gdata_pool,
+ len);
+ if (!event_entry.gdata) {
+ pr_warn(GHES_PFX "ghes gdata pool alloc fail\n");
+ break;
+ }
+
+ memcpy(event_entry.gdata, gdata, len);
+ event_entry.error_severity = sev;
+
+ if (kfifo_in_spinlocked(&ghes_event_ring, &event_entry, 1,
+ &ghes_event_ring_lock))
+ schedule_work(&ghes_event_work);
+ else {
+ pr_warn(GHES_PFX "ghes event queue full\n");
+ break;
+ }
+
if (guid_equal(sec_type, &CPER_SEC_PLATFORM_MEM)) {
struct cper_sec_mem_err *mem_err = acpi_hest_get_payload(gdata);
@@ -1346,6 +1467,12 @@ static int __init ghes_init(void)
else
pr_info(GHES_PFX "Failed to enable APEI firmware first mode.\n");
+ rc = ghes_gdata_pool_init();
+ if (rc) {
+ pr_warn(GHES_PFX "ghes_gdata_pool_init failed\n");
+ goto err;
+ }
+
return 0;
err:
return rc;
diff --git a/include/acpi/ghes.h b/include/acpi/ghes.h
index e3f1cddb4ac8..a3dd82069069 100644
--- a/include/acpi/ghes.h
+++ b/include/acpi/ghes.h
@@ -50,6 +50,34 @@ enum {
GHES_SEV_PANIC = 0x3,
};
+
+#ifdef CONFIG_ACPI_APEI_GHES
+/**
+ * ghes_register_event_notifier - register an event notifier
+ * for the non-fatal HW errors.
+ * @nb: pointer to the notifier_block structure of the event notifier.
+ *
+ * Return : 0 - SUCCESS, non-zero - FAIL.
+ */
+int ghes_register_event_notifier(struct notifier_block *nb);
+
+/**
+ * ghes_unregister_event_notifier - unregister the previously
+ * registered event notifier.
+ * @nb: pointer to the notifier_block structure of the event notifier.
+ */
+void ghes_unregister_event_notifier(struct notifier_block *nb);
+#else
+static inline int ghes_register_event_notifier(struct notifier_block *nb)
+{
+ return -ENODEV;
+}
+
+static inline void ghes_unregister_event_notifier(struct notifier_block *nb)
+{
+}
+#endif
+
int ghes_estatus_pool_init(int num_ghes);
/* From drivers/edac/ghes_edac.c */
--
2.17.1
Add callback function for the non-standard HW errors to the GHES notifier.
Signed-off-by: Shiju Jose <[email protected]>
---
drivers/acpi/apei/ghes.c | 56 ++++++++++++++++++++++++++--------------
1 file changed, 37 insertions(+), 19 deletions(-)
diff --git a/drivers/acpi/apei/ghes.c b/drivers/acpi/apei/ghes.c
index 3b89c7621a0d..0c27ea8ea943 100644
--- a/drivers/acpi/apei/ghes.c
+++ b/drivers/acpi/apei/ghes.c
@@ -578,6 +578,34 @@ static int ghes_handle_arm_hw_error(struct notifier_block *nb,
return NOTIFY_STOP;
}
+static int ghes_handle_non_standard_event(struct notifier_block *nb,
+ unsigned long event, void *data)
+{
+ struct acpi_hest_generic_data *gdata = data;
+ void *err = acpi_hest_get_payload(gdata);
+ int sec_sev;
+ guid_t *sec_type;
+ const guid_t *fru_id = &guid_null;
+ char *fru_text = "";
+
+ if (guid_equal((guid_t *)gdata->section_type, &CPER_SEC_PLATFORM_MEM) ||
+ guid_equal((guid_t *)gdata->section_type, &CPER_SEC_PCIE) ||
+ guid_equal((guid_t *)gdata->section_type, &CPER_SEC_PROC_ARM))
+ return NOTIFY_DONE;
+
+ sec_type = (guid_t *)gdata->section_type;
+ sec_sev = ghes_severity(gdata->error_severity);
+ if (gdata->validation_bits & CPER_SEC_VALID_FRU_ID)
+ fru_id = (guid_t *)gdata->fru_id;
+
+ if (gdata->validation_bits & CPER_SEC_VALID_FRU_TEXT)
+ fru_text = gdata->fru_text;
+
+ log_non_standard_event(sec_type, fru_id, fru_text, sec_sev,
+ err, gdata->error_data_length);
+ return NOTIFY_OK;
+}
+
static struct notifier_block ghes_notifier_mem_error = {
.notifier_call = ghes_handle_memory_failure,
};
@@ -590,6 +618,10 @@ static struct notifier_block ghes_notifier_arm_hw_error = {
.notifier_call = ghes_handle_arm_hw_error,
};
+static struct notifier_block ghes_notifier_non_standard_event = {
+ .notifier_call = ghes_handle_non_standard_event,
+};
+
struct ghes_error_handler_list {
const char *name;
struct notifier_block *nb;
@@ -608,6 +640,10 @@ static const struct ghes_error_handler_list ghes_error_handler_list[] = {
.name = "ghes_notifier_arm_hw_error",
.nb = &ghes_notifier_arm_hw_error,
},
+ {
+ .name = "ghes_notifier_non_standard_event",
+ .nb = &ghes_notifier_non_standard_event,
+ },
};
static BLOCKING_NOTIFIER_HEAD(ghes_event_notify_list);
@@ -656,24 +692,13 @@ static DECLARE_WORK(ghes_event_work, ghes_event_work_func);
static void ghes_do_proc(struct ghes *ghes,
const struct acpi_hest_generic_status *estatus)
{
- int sev, sec_sev;
+ int sev;
struct acpi_hest_generic_data *gdata;
- guid_t *sec_type;
- const guid_t *fru_id = &guid_null;
- char *fru_text = "";
struct ghes_event_entry event_entry;
u32 len;
sev = ghes_severity(estatus->error_severity);
apei_estatus_for_each_section(estatus, gdata) {
- sec_type = (guid_t *)gdata->section_type;
- sec_sev = ghes_severity(gdata->error_severity);
- if (gdata->validation_bits & CPER_SEC_VALID_FRU_ID)
- fru_id = (guid_t *)gdata->fru_id;
-
- if (gdata->validation_bits & CPER_SEC_VALID_FRU_TEXT)
- fru_text = gdata->fru_text;
-
len = acpi_hest_get_record_size(gdata);
event_entry.gdata = (void *)gen_pool_alloc(ghes_gdata_pool,
len);
@@ -692,13 +717,6 @@ static void ghes_do_proc(struct ghes *ghes,
pr_warn(GHES_PFX "ghes event queue full\n");
break;
}
- {
- void *err = acpi_hest_get_payload(gdata);
-
- log_non_standard_event(sec_type, fru_id, fru_text,
- sec_sev, err,
- gdata->error_data_length);
- }
}
}
--
2.17.1
From: Yicong Yang <[email protected]>
The HiSilicon HIP PCIe controller is capable of handling errors
on root port and perform port reset separately at each root port.
Add error handling driver for HIP PCIe controller to log
and report recoverable errors. Perform root port reset and restore
link status after the recovery.
Following are some of the PCIe controller's recoverable errors
1. completion transmission timeout error.
2. CRS retry counter over the threshold error.
3. ECC 2 bit errors
4. AXI bresponse/rresponse errors etc.
The driver placed in the drivers/pci/controller/ because the
HIP PCIe controller does not use DWC ip.
Signed-off-by: Yicong Yang <[email protected]>
Signed-off-by: Shiju Jose <[email protected]>
--
drivers/pci/controller/Kconfig | 8 +
drivers/pci/controller/Makefile | 1 +
drivers/pci/controller/pcie-hisi-error.c | 336 +++++++++++++++++++++++++++++++
3 files changed, 345 insertions(+)
create mode 100644 drivers/pci/controller/pcie-hisi-error.c
---
drivers/pci/controller/Kconfig | 8 +
drivers/pci/controller/Makefile | 1 +
drivers/pci/controller/pcie-hisi-error.c | 323 +++++++++++++++++++++++
3 files changed, 332 insertions(+)
create mode 100644 drivers/pci/controller/pcie-hisi-error.c
diff --git a/drivers/pci/controller/Kconfig b/drivers/pci/controller/Kconfig
index 20bf00f587bd..8bc6111480c8 100644
--- a/drivers/pci/controller/Kconfig
+++ b/drivers/pci/controller/Kconfig
@@ -268,6 +268,14 @@ config PCI_HYPERV_INTERFACE
The Hyper-V PCI Interface is a helper driver allows other drivers to
have a common interface with the Hyper-V PCI frontend driver.
+config PCIE_HISI_ERR
+ depends on ARM64 || COMPILE_TEST
+ depends on ACPI
+ bool "HiSilicon HIP PCIe controller error handling driver"
+ help
+ Say Y here if you want error handling support
+ for the PCIe controller's errors on HiSilicon HIP SoCs
+
source "drivers/pci/controller/dwc/Kconfig"
source "drivers/pci/controller/cadence/Kconfig"
endmenu
diff --git a/drivers/pci/controller/Makefile b/drivers/pci/controller/Makefile
index 01b2502a5323..94f37b3d9929 100644
--- a/drivers/pci/controller/Makefile
+++ b/drivers/pci/controller/Makefile
@@ -29,6 +29,7 @@ obj-$(CONFIG_PCIE_MOBIVEIL) += pcie-mobiveil.o
obj-$(CONFIG_PCIE_TANGO_SMP8759) += pcie-tango.o
obj-$(CONFIG_VMD) += vmd.o
obj-$(CONFIG_PCIE_BRCMSTB) += pcie-brcmstb.o
+obj-$(CONFIG_PCIE_HISI_ERR) += pcie-hisi-error.o
# pcie-hisi.o quirks are needed even without CONFIG_PCIE_DW
obj-y += dwc/
diff --git a/drivers/pci/controller/pcie-hisi-error.c b/drivers/pci/controller/pcie-hisi-error.c
new file mode 100644
index 000000000000..cc721070e07b
--- /dev/null
+++ b/drivers/pci/controller/pcie-hisi-error.c
@@ -0,0 +1,323 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Driver for handling the PCIe controller errors on
+ * HiSilicon HIP SoCs.
+ *
+ * Copyright (c) 2018-2019 HiSilicon Limited.
+ */
+
+#include <linux/acpi.h>
+#include <acpi/ghes.h>
+#include <linux/delay.h>
+#include <linux/pci.h>
+#include <linux/platform_device.h>
+#include <linux/kfifo.h>
+#include <linux/spinlock.h>
+
+#define HISI_PCIE_ERR_INFO_SIZE 1024
+
+/* HISI PCIe controller error definitions */
+#define HISI_PCIE_ERR_MISC_REGS 33
+
+#define HISI_PCIE_SUB_MODULE_ID_AP 0
+#define HISI_PCIE_SUB_MODULE_ID_TL 1
+#define HISI_PCIE_SUB_MODULE_ID_MAC 2
+#define HISI_PCIE_SUB_MODULE_ID_DL 3
+#define HISI_PCIE_SUB_MODULE_ID_SDI 4
+
+#define HISI_PCIE_LOCAL_VALID_VERSION BIT(0)
+#define HISI_PCIE_LOCAL_VALID_SOC_ID BIT(1)
+#define HISI_PCIE_LOCAL_VALID_SOCKET_ID BIT(2)
+#define HISI_PCIE_LOCAL_VALID_NIMBUS_ID BIT(3)
+#define HISI_PCIE_LOCAL_VALID_SUB_MODULE_ID BIT(4)
+#define HISI_PCIE_LOCAL_VALID_CORE_ID BIT(5)
+#define HISI_PCIE_LOCAL_VALID_PORT_ID BIT(6)
+#define HISI_PCIE_LOCAL_VALID_ERR_TYPE BIT(7)
+#define HISI_PCIE_LOCAL_VALID_ERR_SEVERITY BIT(8)
+#define HISI_PCIE_LOCAL_VALID_ERR_MISC 9
+
+#define HISI_ERR_SEV_RECOVERABLE 0
+#define HISI_ERR_SEV_FATAL 1
+#define HISI_ERR_SEV_CORRECTED 2
+#define HISI_ERR_SEV_NONE 3
+
+static guid_t hisi_pcie_sec_type = GUID_INIT(0xB2889FC9, 0xE7D7, 0x4F9D,
+ 0xA8, 0x67, 0xAF, 0x42, 0xE9, 0x8B, 0xE7, 0x72);
+
+#define HISI_PCIE_CORE_ID(v) ((v) >> 3)
+#define HISI_PCIE_PORT_ID(core, v) (((v) >> 1) + ((core) << 3))
+#define HISI_PCIE_CORE_PORT_ID(v) (((v) % 8) << 1)
+
+struct hisi_pcie_error_data {
+ u64 val_bits;
+ u8 version;
+ u8 soc_id;
+ u8 socket_id;
+ u8 nimbus_id;
+ u8 sub_module_id;
+ u8 core_id;
+ u8 port_id;
+ u8 err_severity;
+ u16 err_type;
+ u8 reserv[2];
+ u32 err_misc[HISI_PCIE_ERR_MISC_REGS];
+};
+
+struct hisi_pcie_error_private {
+ struct notifier_block nb;
+ struct platform_device *pdev;
+};
+
+static char *hisi_pcie_sub_module_name(u8 id)
+{
+ switch (id) {
+ case HISI_PCIE_SUB_MODULE_ID_AP: return "AP Layer";
+ case HISI_PCIE_SUB_MODULE_ID_TL: return "TL Layer";
+ case HISI_PCIE_SUB_MODULE_ID_MAC: return "MAC Layer";
+ case HISI_PCIE_SUB_MODULE_ID_DL: return "DL Layer";
+ case HISI_PCIE_SUB_MODULE_ID_SDI: return "SDI Layer";
+ }
+
+ return "unknown";
+}
+
+static char *hisi_pcie_error_severity(u8 err_sev)
+{
+ switch (err_sev) {
+ case HISI_ERR_SEV_RECOVERABLE: return "recoverable";
+ case HISI_ERR_SEV_FATAL: return "fatal";
+ case HISI_ERR_SEV_CORRECTED: return "corrected";
+ case HISI_ERR_SEV_NONE: return "none";
+ }
+
+ return "unknown";
+}
+
+static int hisi_pcie_port_reset(struct platform_device *pdev,
+ u32 chip_id, u32 port_id)
+{
+ struct device *dev = &pdev->dev;
+ acpi_handle handle = ACPI_HANDLE(dev);
+ union acpi_object arg[3];
+ struct acpi_object_list arg_list;
+ acpi_status s;
+ unsigned long long data = 0;
+
+ arg[0].type = ACPI_TYPE_INTEGER;
+ arg[0].integer.value = chip_id;
+ arg[1].type = ACPI_TYPE_INTEGER;
+ arg[1].integer.value = HISI_PCIE_CORE_ID(port_id);
+ arg[2].type = ACPI_TYPE_INTEGER;
+ arg[2].integer.value = HISI_PCIE_CORE_PORT_ID(port_id);
+
+ arg_list.count = 3;
+ arg_list.pointer = arg;
+
+ s = acpi_evaluate_integer(handle, "RST", &arg_list, &data);
+ if (ACPI_FAILURE(s)) {
+ dev_err(dev, "No RST method\n");
+ return -EIO;
+ }
+
+ if (data) {
+ dev_err(dev, "Failed to Reset\n");
+ return -EIO;
+ }
+
+ return 0;
+}
+
+static int hisi_pcie_port_do_recovery(struct platform_device *dev,
+ u32 chip_id, u32 port_id)
+{
+ acpi_status s;
+ struct device *device = &dev->dev;
+ acpi_handle root_handle = ACPI_HANDLE(device);
+ struct acpi_pci_root *pci_root;
+ struct pci_bus *root_bus;
+ struct pci_dev *pdev;
+ u32 domain, busnr, devfn;
+
+ s = acpi_get_parent(root_handle, &root_handle);
+ if (ACPI_FAILURE(s))
+ return -ENODEV;
+ pci_root = acpi_pci_find_root(root_handle);
+ if (!pci_root)
+ return -ENODEV;
+ root_bus = pci_root->bus;
+ domain = pci_root->segment;
+
+ busnr = root_bus->number;
+ devfn = PCI_DEVFN(port_id, 0);
+ pdev = pci_get_domain_bus_and_slot(domain, busnr, devfn);
+ if (!pdev) {
+ dev_info(device, "Fail to get root port %04x:%02x:%02x.%d device\n",
+ domain, busnr, PCI_SLOT(devfn), PCI_FUNC(devfn));
+ return -ENODEV;
+ }
+
+ pci_stop_and_remove_bus_device_locked(pdev);
+ pci_dev_put(pdev);
+
+ if (hisi_pcie_port_reset(dev, chip_id, port_id))
+ return -EIO;
+
+ /*
+ * The initialization time of subordinate devices after
+ * hot reset is no more than 1s, which is required by
+ * the PCI spec v5.0 sec 6.6.1. The time will shorten
+ * if Readiness Notifications mechanisms are used. But
+ * wait 1s here to adapt any conditions.
+ */
+ ssleep(1UL);
+
+ /* add root port and downstream devices */
+ pci_lock_rescan_remove();
+ pci_rescan_bus(root_bus);
+ pci_unlock_rescan_remove();
+
+ return 0;
+}
+
+static void hisi_pcie_handle_error(const struct hisi_pcie_error_data *error,
+ struct platform_device *pdev)
+{
+ char buf[HISI_PCIE_ERR_INFO_SIZE];
+ char *p = buf, *end = buf + sizeof(buf);
+ struct device *dev = &pdev->dev;
+ u32 i;
+ int rc;
+
+ if (error->val_bits == 0) {
+ dev_warn(dev, "%s: no valid error information\n", __func__);
+ return;
+ }
+
+ /* Logging */
+ p += snprintf(p, end - p, "[ Table version=%d ", error->version);
+ if (error->val_bits & HISI_PCIE_LOCAL_VALID_SOC_ID)
+ p += snprintf(p, end - p, "SOC ID=%d ", error->soc_id);
+
+ if (error->val_bits & HISI_PCIE_LOCAL_VALID_SOCKET_ID)
+ p += snprintf(p, end - p, "socket ID=%d ", error->socket_id);
+
+ if (error->val_bits & HISI_PCIE_LOCAL_VALID_NIMBUS_ID)
+ p += snprintf(p, end - p, "nimbus ID=%d ", error->nimbus_id);
+
+ if (error->val_bits & HISI_PCIE_LOCAL_VALID_SUB_MODULE_ID)
+ p += snprintf(p, end - p, "sub module=%s ",
+ hisi_pcie_sub_module_name(error->sub_module_id));
+
+ if (error->val_bits & HISI_PCIE_LOCAL_VALID_CORE_ID)
+ p += snprintf(p, end - p, "core ID=core%d ", error->core_id);
+
+ if (error->val_bits & HISI_PCIE_LOCAL_VALID_PORT_ID)
+ p += snprintf(p, end - p, "port ID=port%d ", error->port_id);
+
+ if (error->val_bits & HISI_PCIE_LOCAL_VALID_ERR_SEVERITY)
+ p += snprintf(p, end - p, "error severity=%s ",
+ hisi_pcie_error_severity(error->err_severity));
+
+ if (error->val_bits & HISI_PCIE_LOCAL_VALID_ERR_TYPE)
+ p += snprintf(p, end - p, "error type=0x%x ", error->err_type);
+
+ p += snprintf(p, end - p, "]\n");
+ dev_info(dev, "\nHISI : HIP : PCIe controller error\n");
+ dev_info(dev, "%s\n", buf);
+
+ dev_info(dev, "Reg Dump:\n");
+ for (i = 0; i < HISI_PCIE_ERR_MISC_REGS; i++) {
+ if (error->val_bits &
+ BIT_ULL(HISI_PCIE_LOCAL_VALID_ERR_MISC + i))
+ dev_info(dev,
+ "ERR_MISC_%d=0x%x\n", i, error->err_misc[i]);
+ }
+
+ /* Recovery for the PCIe controller errors */
+ if (error->err_severity == HISI_ERR_SEV_RECOVERABLE) {
+ /* try reset PCI port for the error recovery */
+ rc = hisi_pcie_port_do_recovery(pdev, error->socket_id,
+ HISI_PCIE_PORT_ID(error->core_id, error->port_id));
+ if (rc) {
+ dev_info(dev, "fail to do hisi pcie port reset\n");
+ return;
+ }
+ }
+}
+
+static int hisi_pcie_notify_error(struct notifier_block *nb,
+ unsigned long event, void *data)
+{
+ struct acpi_hest_generic_data *gdata = data;
+ const struct hisi_pcie_error_data *error_data =
+ acpi_hest_get_payload(gdata);
+ struct hisi_pcie_error_private *priv =
+ container_of(nb, struct hisi_pcie_error_private, nb);
+ struct platform_device *pdev = priv->pdev;
+ struct device *dev = &pdev->dev;
+ u8 socket;
+
+ if (device_property_read_u8(dev, "socket", &socket))
+ return NOTIFY_DONE;
+
+ if (!guid_equal((guid_t *)gdata->section_type, &hisi_pcie_sec_type) ||
+ error_data->socket_id != socket)
+ return NOTIFY_DONE;
+
+ hisi_pcie_handle_error(error_data, pdev);
+
+ return NOTIFY_OK;
+}
+
+static int hisi_pcie_error_handler_probe(struct platform_device *pdev)
+{
+ struct hisi_pcie_error_private *priv;
+ int ret;
+
+ priv = devm_kzalloc(&pdev->dev, sizeof(*priv), GFP_KERNEL);
+ if (!priv)
+ return -ENOMEM;
+
+ priv->nb.notifier_call = hisi_pcie_notify_error;
+ priv->pdev = pdev;
+ ret = ghes_register_event_notifier(&priv->nb);
+ if (ret) {
+ dev_err(&pdev->dev, "%s : ghes_register_event_notifier fail\n",
+ __func__);
+ return ret;
+ }
+
+ platform_set_drvdata(pdev, priv);
+
+ return 0;
+}
+
+static int hisi_pcie_error_handler_remove(struct platform_device *pdev)
+{
+ struct hisi_pcie_error_private *priv = platform_get_drvdata(pdev);
+
+ if (priv)
+ ghes_unregister_event_notifier(&priv->nb);
+
+ kfree(priv);
+
+ return 0;
+}
+
+static const struct acpi_device_id hisi_pcie_acpi_match[] = {
+ { "HISI0361", 0 },
+ { }
+};
+
+static struct platform_driver hisi_pcie_error_handler_driver = {
+ .driver = {
+ .name = "hisi-pcie-error-handler",
+ .acpi_match_table = hisi_pcie_acpi_match,
+ },
+ .probe = hisi_pcie_error_handler_probe,
+ .remove = hisi_pcie_error_handler_remove,
+};
+module_platform_driver(hisi_pcie_error_handler_driver);
+
+MODULE_DESCRIPTION("HiSilicon HIP PCIe controller error handling driver");
+MODULE_LICENSE("GPL v2");
--
2.17.1
Hi Shiju,
Thank you for the patch! Perhaps something to improve:
[auto build test WARNING on v5.6]
[also build test WARNING on next-20200407]
[cannot apply to pm/linux-next pci/next linus/master linux/master]
[if your patch is applied to the wrong git tree, please drop us a note to help
improve the system. BTW, we also suggest to use '--base' option to specify the
base tree in git format-patch, please see https://stackoverflow.com/a/37406982]
url: https://github.com/0day-ci/linux/commits/Shiju-Jose/ACPI-APEI-Add-support-to-notify-non-fatal-HW-errors/20200408-014447
base: 7111951b8d4973bda27ff663f2cf18b663d15b48
config: i386-allyesconfig (attached as .config)
compiler: gcc-7 (Ubuntu 7.5.0-6ubuntu2) 7.5.0
reproduce:
# save the attached .config to linux build tree
make ARCH=i386
If you fix the issue, kindly add following tag as appropriate
Reported-by: kbuild test robot <[email protected]>
All warnings (new ones prefixed by >>):
drivers/pci/controller/pcie-hisi-error.c: In function 'hisi_pcie_handle_error':
>> drivers/pci/controller/pcie-hisi-error.c:246:1: warning: the frame size of 1124 bytes is larger than 1024 bytes [-Wframe-larger-than=]
}
^
vim +246 drivers/pci/controller/pcie-hisi-error.c
181
182 static void hisi_pcie_handle_error(const struct hisi_pcie_error_data *error,
183 struct platform_device *pdev)
184 {
185 char buf[HISI_PCIE_ERR_INFO_SIZE];
186 char *p = buf, *end = buf + sizeof(buf);
187 struct device *dev = &pdev->dev;
188 u32 i;
189 int rc;
190
191 if (error->val_bits == 0) {
192 dev_warn(dev, "%s: no valid error information\n", __func__);
193 return;
194 }
195
196 /* Logging */
197 p += snprintf(p, end - p, "[ Table version=%d ", error->version);
198 if (error->val_bits & HISI_PCIE_LOCAL_VALID_SOC_ID)
199 p += snprintf(p, end - p, "SOC ID=%d ", error->soc_id);
200
201 if (error->val_bits & HISI_PCIE_LOCAL_VALID_SOCKET_ID)
202 p += snprintf(p, end - p, "socket ID=%d ", error->socket_id);
203
204 if (error->val_bits & HISI_PCIE_LOCAL_VALID_NIMBUS_ID)
205 p += snprintf(p, end - p, "nimbus ID=%d ", error->nimbus_id);
206
207 if (error->val_bits & HISI_PCIE_LOCAL_VALID_SUB_MODULE_ID)
208 p += snprintf(p, end - p, "sub module=%s ",
209 hisi_pcie_sub_module_name(error->sub_module_id));
210
211 if (error->val_bits & HISI_PCIE_LOCAL_VALID_CORE_ID)
212 p += snprintf(p, end - p, "core ID=core%d ", error->core_id);
213
214 if (error->val_bits & HISI_PCIE_LOCAL_VALID_PORT_ID)
215 p += snprintf(p, end - p, "port ID=port%d ", error->port_id);
216
217 if (error->val_bits & HISI_PCIE_LOCAL_VALID_ERR_SEVERITY)
218 p += snprintf(p, end - p, "error severity=%s ",
219 hisi_pcie_error_severity(error->err_severity));
220
221 if (error->val_bits & HISI_PCIE_LOCAL_VALID_ERR_TYPE)
222 p += snprintf(p, end - p, "error type=0x%x ", error->err_type);
223
224 p += snprintf(p, end - p, "]\n");
225 dev_info(dev, "\nHISI : HIP : PCIe controller error\n");
226 dev_info(dev, "%s\n", buf);
227
228 dev_info(dev, "Reg Dump:\n");
229 for (i = 0; i < HISI_PCIE_ERR_MISC_REGS; i++) {
230 if (error->val_bits &
231 BIT_ULL(HISI_PCIE_LOCAL_VALID_ERR_MISC + i))
232 dev_info(dev,
233 "ERR_MISC_%d=0x%x\n", i, error->err_misc[i]);
234 }
235
236 /* Recovery for the PCIe controller errors */
237 if (error->err_severity == HISI_ERR_SEV_RECOVERABLE) {
238 /* try reset PCI port for the error recovery */
239 rc = hisi_pcie_port_do_recovery(pdev, error->socket_id,
240 HISI_PCIE_PORT_ID(error->core_id, error->port_id));
241 if (rc) {
242 dev_info(dev, "fail to do hisi pcie port reset\n");
243 return;
244 }
245 }
> 246 }
247
---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/[email protected]
Signed-off-by: kbuild test robot <[email protected]>
---
ghes.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/acpi/apei/ghes.c b/drivers/acpi/apei/ghes.c
index 5c0ab54223118..6d698867468d1 100644
--- a/drivers/acpi/apei/ghes.c
+++ b/drivers/acpi/apei/ghes.c
@@ -207,7 +207,7 @@ int ghes_estatus_pool_init(int num_ghes)
return -ENOMEM;
}
-int ghes_gdata_pool_init(void)
+static int ghes_gdata_pool_init(void)
{
unsigned long addr, len;
int rc;
Hi Shiju,
Thank you for the patch! Perhaps something to improve:
[auto build test WARNING on v5.6]
[also build test WARNING on next-20200408]
[cannot apply to pm/linux-next pci/next linus/master linux/master]
[if your patch is applied to the wrong git tree, please drop us a note to help
improve the system. BTW, we also suggest to use '--base' option to specify the
base tree in git format-patch, please see https://stackoverflow.com/a/37406982]
url: https://github.com/0day-ci/linux/commits/Shiju-Jose/ACPI-APEI-Add-support-to-notify-non-fatal-HW-errors/20200408-014447
base: 7111951b8d4973bda27ff663f2cf18b663d15b48
reproduce:
# apt-get install sparse
# sparse version: v0.6.1-188-g79f7ac98-dirty
make ARCH=x86_64 allmodconfig
make C=1 CF='-fdiagnostic-prefix -D__CHECK_ENDIAN__'
If you fix the issue, kindly add following tag as appropriate
Reported-by: kbuild test robot <[email protected]>
sparse warnings: (new ones prefixed by >>)
>> drivers/acpi/apei/ghes.c:210:5: sparse: sparse: symbol 'ghes_gdata_pool_init' was not declared. Should it be static?
drivers/acpi/apei/ghes.c:715:25: sparse: sparse: incompatible types in comparison expression (different address spaces):
drivers/acpi/apei/ghes.c:715:25: sparse: struct ghes_estatus_cache [noderef] <asn:4> *
drivers/acpi/apei/ghes.c:715:25: sparse: struct ghes_estatus_cache *
drivers/acpi/apei/ghes.c:795:25: sparse: sparse: incompatible types in comparison expression (different address spaces):
drivers/acpi/apei/ghes.c:795:25: sparse: struct ghes_estatus_cache [noderef] <asn:4> *
drivers/acpi/apei/ghes.c:795:25: sparse: struct ghes_estatus_cache *
Please review and possibly fold the followup patch.
---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/[email protected]